Home

Non-deterministic results often arise unexpectedly in High Performance Computing (HPC) applications. These make reproducible and correct results difficult to create. As such, there is a need to improve the ability of software developers and scientists to comprehend root sources of non-determinism in their applications. To this end, we present ANACIN-X. This wiki is organized into the following pages:

• An Overview of the Software.
• Software Installation Please read this section carefully as there are multiple steps involved.
• Software Dependencies for ANACIN-X We do not recommend installing all of the dependencies on your own. Instead, use instructions in the installation section to install the dependencies, starting with those on installing Spack and Conda.
• Running ANACIN-X. This is broken up into 2 main parts:
  1. Running ANACIN-X on a Benchmark Application
  2. Running ANACIN-X on an External Application
• Visualization This is broken up into 2 main parts:
  1. Visualizing Benchmark Application Data
  2. Visualizing External Application Data
• Reproducing Published Results
• Additional Project Details
  1. Supported systems
  2. The project team
  3. Publications associated with the software
  4. Copyright and license information
• Ready-to-use Apptainer container. This container presents all the ANACIN-X software packages to test the included benchmarks and visualize their results.